Which Work Stays Human

The 2027 AI Action Plan for Small Business · Post 2 of 8 · See the full series

Look at any role in your company. For the entire history of running a business, every job required two things to happen inside the same person: the thinking and the doing.

The thinking part is the judgment. That is your accountant understanding what a tax variance actually means for your cash flow, or your estimator knowing a specific client will reject a certain material. The doing part is the production. That is the exact same accountant spending three hours pulling data from different software to find the variance, or the estimator spending an hour manually formatting the proposal document.

You have always paid for both of those things bundled together. That was never a strategic choice. It was just a physical constraint, because there was no way to separate the judgment from the production.

AI breaks that bundle. For the first time, you can separate the two halves.

When small business owners ask me if they should use AI to replace a specific job title, they are asking the wrong question. Job titles are just bundles of tasks. The right question is much narrower: when you look at a specific task, which half is production that a machine can handle, and which half is the irreplaceable human judgment you actually need?

The Uncomfortable Truth About Expert Work

In Workforce Intelligence, Mike Sullivan and Vinay Gidwaney outline what OneDigital found when the firm ran a large internal disruption assessment. The results challenged some core assumptions about what counts as high-level work:

“It revealed that some of the work we thought was most sophisticated was actually quite reducible. Pattern matching against large datasets. Regulatory lookups. Comparative benchmarking. Structured research synthesis.”

Workforce Intelligence, Chapter 11

For a small business owner, that finding hits close to home. Some of the work you take the most pride in is highly reducible. Regulatory lookups, benchmarking competitors, and drafting complex reports feel like expert work because they used to require an expert. They still require your judgment to direct. They simply no longer require your time to produce.

Saying this plainly is more useful than softening it. The owners who act on this distinction gain time and a real competitive advantage. The owners who protect those production tasks as irreplaceable lose both.

The Irreducibility Test for Small Businesses

To figure out exactly where AI belongs in your operations, you need to run an irreducibility test. A large enterprise might formally test a hundred different functions. A small business only needs to map out six to ten core work centers and spend an afternoon analyzing them.

For each core function, separate the work into two buckets:

  • Judgment work. This requires your context, your relationships, your legal accountability, and your read of a situation. It cannot be delegated to AI without degrading the output in ways your clients would immediately notice.
  • Production work. This includes the drafting, formatting, calculating, benchmarking, and summarizing. AI handles this well with the right direction.

Once you separate the judgment from the production, every task in your business falls into one of four categories:

CategoryWhat it isWho does the workWhere to start
EliminateHigh-volume, rule-based work with clear right answersAI handles it end to endUseful, but rarely where your most strategic time goes
RedesignWork where AI drafts and analyzes, and you direct, review and decideAI does the production; you keep the judgmentStart here. This is the richest category
RetainWork where the human element is the core valueYouDo not spend optimization energy here first
CreateNew capabilities that AI makes possibleAI and your team togetherLast. Fund it with the capacity Redesign returns

Eliminate

This is high-volume, rule-based work with clear right answers. AI handles this end to end. This is rarely where your true value lies, because fully eliminable tasks rarely consume your most strategic time.

Redesign

This is the richest category and exactly where you should start. AI drafts and analyzes, while you direct, review, and decide. Take client renewals as an example. The background research, the comparative analysis, and the first draft of the recommendation are all production work. AI handles that. The renewal conversation stays yours. You walk into the meeting better prepared and able to give more time to the judgment part.

Retain

The human element is the core value here. This includes the termination conversation, the relationship with a key client built over years, or the judgment call in a complex compliance situation where the right answer depends on context you carry in your head. AI adds very little here. The risk is spending optimization energy on these tasks when redesign tasks are waiting.

Create

These are entirely new capabilities made possible by AI. This might be proactive market analysis that used to take a week but now takes an afternoon. This is the most exciting category, but it is the worst place to start. Get your redesigned work running smoothly first. The capacity it returns is what funds the new creations.

What This Looks Like in Practice

When you map this out for a small business, the patterns become obvious.

TaskProduction work for AIJudgment work that stays humanCategory
Month-end closeData pull and reconciliationReview and sign-offRedesign
Quoting a jobMaterials lookup and labor calculationScope assessment and final margin callRedesign
Collections callsAccount summary and payment historyRelationship risk and the negotiation itselfRedesign

Notice what these have in common. Your time shifts toward the part that was always the most valuable, and AI handles the part that was consuming the most hours. The output does not get worse. It gets better because you arrive fully prepared.

Your Next Steps

I strongly recommend downloading the free 2027 AI Transformation Worksheet. It is a simple, printable two-page worksheet designed to help you sit down with a pen and map out your core work centers. It makes you write down which parts require human judgment and which parts are just habit.

Before you commit to a new workflow, take two minutes to run the project through the AI Opportunity Scorecard. It gives you a verdict on whether a specific project is the right first move, plus a custom action plan telling you exactly what to fix first.

If you want to talk through where your specific operations sit in this analysis, the Focus HR team works through exactly this with Arizona small business owners every day.

Talk to the Focus HR team

About the author: Clint Parry, MBA, SHRM-SCP is a Senior Business Consultant at Focus HR, where he helps Arizona small businesses with HR strategy, compliance and workforce planning. Focus HR is powered by OneDigital. This post is part of The 2027 AI Action Plan for Small Business series.

Leave a Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Scroll to Top